Lakera builds runtime security for large language models and generative AI systems. The tech stack—Python, C++, Go, React, Kubernetes, Terraform—reflects a systems-engineering foundation typical of infrastructure security. Active projects center on scaling red-teaming automation and foundation model defenses, while pain points cluster around adversarial attack mitigation and customer scaling, suggesting the company is transitioning from product-market fit toward operationalizing both technical depth (security research) and go-to-market (sales velocity, customer success).
Notable leadership hires: Red Team Lead
Lakera is a San Francisco-based GenAI security company founded in 2021. The platform detects and blocks adversarial prompts and misuse in real time without degrading application performance. Beyond the core product, Lakera operates Gandalf, an educational platform where over one million users have engaged with AI security concepts. The company serves engineering and security teams at organizations adopting large language models, with a 50/50 split between engineering and commercial hiring, and active recruiting across nine countries including the U.S., UK, Switzerland, and Asia-Pacific.
Python, C++, Go for backend; React and TypeScript for frontend. Infrastructure: AWS, GCP, Azure, Kubernetes, Terraform, Docker. Observability: Datadog, Grafana, Sentry. Currently adopting containerization and infrastructure-as-code tooling (Docker, Kubernetes, Terraform).
Scaling foundation model defenses and red-teaming automation, building repeatable playbooks and best practices, designing an AI defense plane architecture, and establishing a scalable solutions engineering function to support customer onboarding and configuration.
Other companies in the same industry, closest in size